MaxMine's machine learning system cuts mining workload, improves accuracy after six months in production
MaxMine has deployed a production-grade machine learning system for load and dump classification across major Australian mining operations, including sites run by Glencore, NRW Holdings, and Macmahon. The system has been operational for six months.
The ML system reduces site team workload by catching missed or incorrect loads. It also improves production tracking accuracy in complex scenarios and handles edge cases without requiring custom development for each situation.
What's driving the results
The system draws on over 14 million hours of labeled operational data and high-resolution datasets from load and haul operations. This foundation lets the ML model deliver consistent, high-confidence results in live production environments.
Shaun Mitchell, CEO at MaxMine, said the success reflects a broader pattern: "Organisations succeeding in AI are those that have the highest-quality datasets."
Professor Anton Van Den Hengel, Chief Scientist at the Australian Institute of Machine Learning, added that deploying a model with this accuracy across different asset types and sites is uncommon. "MaxMine's ability to do this points to their uniquely rich, accurate, and human error-free data sets, paired with long-term, multi-site, multi-machine training data sets."
The broader challenge
Most AI projects in mining and other industrial sectors struggle to move beyond pilots to operational scale. Gartner estimates that 60% of AI projects fail due to lack of AI-ready data, and 42% of organisations abandon AI initiatives before reaching production.
Tom Cawley, Executive Chair at MaxMine, was recently appointed mining sector lead for the newly established AI Accelerator Cooperative Research Centre. The centre aims to build Australia's sovereign AI creation capacity so sectors like mining can develop their own tools rather than relying on offshore capabilities.
Cawley said: "At MaxMine, we've demonstrated that Australia has the capability to develop advanced AI tools that work effectively at scale in mining. I hope my role at the AI Accelerator CRC will encourage further innovation across the sector."
Operations teams interested in AI for Operations should understand that success depends on the quality and consistency of the data feeding these systems. Data Analysis skills become critical as organisations move AI projects into production.
Your membership also unlocks:
